Node core committer Erick Wendel joins Jerod & KBall to talk us through how he created his own JS runtime using V8, Libuv & more. Along the way we learn from his learnings, wrap our heads around the differences between Node, Bun & Deno, and talk about creating awesome content for developers… whether they like it or not!
Fly.io – The home of Changelog.com — Deploy your apps and databases close to your users. In minutes you can run your Ruby, Go, Node, Deno, Python, or Elixir app (and databases!) all over the world. No ops required. Learn more at fly.io/changelog and check out the speedrun in their docs.
- Holla! BeJS CONF Use code
JSPARTYTIMEfor 30% off
- ECMAScript 2023 Language Specification
|Chapter Number||Chapter Start Time||Chapter Title|
|1||00:00||It's party time, y'all!|
|2||01:03||Welcoming Erick to the show|
|3||03:19||Holla! BeJS CONF|
|4||05:20||Erick's Node.js research|
|5||08:59||Powered by V8|
|6||17:41||Where libuv fits in|
|8||30:55||Sponsor: KBall Growth|
|9||31:38||First steps to recreate Node|
|10||35:23||Drawing clear lines around V8|
|11||44:09||Jerod breaks ChatGPT|
|12||45:05||Erick's curious about React Native|
|13||47:04||Persistence and video games|
|14||49:20||The content creator's dilemma|
|15||51:58||Erick's publishing formula|
|16||54:30||Erick's final words|
|17||55:13||KBall's final words|
|18||56:45||Jerod's final words|
|19||57:25||The wrap up|
|20||58:57||Next up on the pod|
Play the audio to listen along while you enjoy the transcript. 🎧
Hey, hey. Glad to be here.
You, sir, are an independent businessman now, aren’t you? Why don’t you give us a status update? I know on New Year’s you talked about that, and you’ve been doing it a little bit now… Do you wanna tell us real quick what you’re up to on your day-to-day?
Yeah, so I am – and in fact, if you’re listening to JS Party, you may have heard I’m even running an ad.
Talk about businessman…
But no, I’ve been coaching engineers and engineering managers, and it’s been great. It’s really fun to get to talk with folks, and work with them and help them improve… And you know me, I love geeking out with tech people… And here we’re mostly working on like if you’re stuck, how do we work you through a non-technical problem… But it’s just fun. And yeah, I’m working towards a new podcast that I’m gonna be releasing a while in the future, but I’m doing interviews and I’m gonna start putting those out on YouTube…
…and working on all sorts of stuff. So it’s good, but that’s not the focus here…
Yes, we have here with us for the first time - hopefully not the last time - Erick Wendel. Welcome to the show.
I’m excited to learn, because this is an area of expertise that I completely lack, personally. So I’m ready to learn from you, Erick. For those who don’t know him - you are a Node.js core committer, keynote speaker, professional educator, you’ve got a great YouTube channel, a bunch of stuff online, helping everybody learn his wonderful world. And you’re gonna be speaking here soon at BeJS Conf, which leads us into our regular segment, Holla.
Erick, you’re speaking at this conference… Have you been to it before?
Actually, I’ve been to Brussels… The best beer I’ve ever had.
Actually, the biggest, I would say too… So I’m really looking forward to seeing more places. I’ve been there just for a day… But it’s a pretty nice city. The environment, the people… It’s very – I like my country; people are very receptionists People really like other people, right?
Very cool. Well, I’m still excited to go there someday. I know Omar are still working on that behind the scenes. One of these days we will connect those dots and I will be there in the flesh. But if you can’t get there in the flash, you can get there online, definitely do that. I highly recommend it. This has been Holla.
So a lot of those questions I saw people actually complaining on the internet, and I saw a lot of theory about this… But in the end it was like “This doesn’t seem right.”
So I actually started contributing to Node.js, I actually solved a really nice bug there… But I felt like I just wanted to learn more about this tech, and I think I want to give a deep knowledge about this. So I asked actually ChatGPT back then, I was like “How is Node made?” So what do I need to make – actually, I don’t remember if I did the same research for ChatGPT back then, but I for the research now I was doing the same questions.
In the beginning, when you look at the Node.js core and how this stuff was working, actually they were exposing everything. Like “Oh, I want to say –” Like, I’m gonna use the console.log. The console.log - we have the STDOUT to receive a data, or to actually throw some error. We have the std error. So we were exposing these functions; we have like a C++ call that received that variable, and we were printing that.
This sounds a bit confusing when you just hear it, but when you’re looking at the code, it’s pretty straightforward. It’s like “Oh, I’m gonna call something in English that is gonna make something easier from their side, but here in Portuguese is more accessible for me to handle this piece of data just using the language I already know, so I don’t have to handle any memory.
But just to conclude about the V8 - in this case, all the optimizations that we have above the language, like for loops, the syntax, and everything, V8 is doing behind the scenes with the V8, with Google. So in my side, as a Node.js contributor, I barely touch the C++ side. I go on the high-level APIs, child processes, or strings, just making this back and forth, but receiving this data from the low level and trying to make the life of users easier.
Okay, so let me see if I understand, because this has kind of blown my mind a little bit.
Just to get some vocabulary - Node.js, Deno and React Native and all this stuff, they are trying to embed V8. So this is one of the pieces of the puzzle. So when you are saying data, [unintelligible 00:15:04.19] but I was following a lot of the Bun and how it actually is affecting Node.js right now… And they say “Oh, we are consuming buffers, we are consuming strings, but if I put here – like, I break this string the middle, I can send it faster to the user land.”
So every runtime now is trying to adapt these algorithms to send data faster, and actually easier to the user, and safer to the runtime as well. This is mind-blowing for me.
But I’ve seen – like, Cloudflare workers I’ve never tried, but I’ve seen a lot of benchmarking people who are saying a lot of good stuff about them… But the worker threads, and a lot of internal stuff would be different, they would implement as they want. This is what I was talking to - oh, in Node.js we use printf, but they could use Ruby, or they could use Rust to bring this differently, so they can implement whatever they want.
Sure, yeah. And it’s pretty hard to embed I don’t know how they did. The docs are very hard.
Right, sure. Okay, so that makes sense. And so you also mentioned another moving part, which is Libuv, which you said allows for asynchronicity. Is Libuv then a separate thing from V8? Is it part of V8? Is it used by these other projects, or do they use something else?
Actually, V8 and Libuv is a library that was made for Node in the beginning.
So the Libuv makes the process easier to make multi-platform of this. So you can call internal APIs for every different operating system from there. I’m talking this because you could actually use only C++ threads if you want, and handle [unintelligible 00:19:16.29] - there’s a lot of crazy other concepts - by hand. But Libuv makes this even easier to make everything.
Yeah. It’s a relatively thin layer. Now, there’s a large API surface with Node and with these other runtimes. So while it’s a thing layer, it’s broad, meaning it has a lot of surface. And so that’s why it’s still a lot of work, right? You might wonder, “Wow, maybe Node isn’t a big deal.” Well, it’s a lot of code, and a lot of surface area.
I did look up while you were talking… It seems that Deno is using Tokyo, which is a Rust package for asyncio, which is one of the differences between Deno and Node… So just trying to google while you were talking so we can –
I know you’re just going from memory. So okay… Kball, do you have thoughts before we move on on this topic?
Yeah, I mean - I think it’s a really interesting point of… So we’ve been talking a lot about the different – there’s just been a lot of buzz in the community that “Oh, Bun is so fast. It’s doing this.” And Deno was pushing this, and doing that. And that sort of mentality - these are places where there’s a lot of competition and differentiation happens… It almost feels a little at odds with the view of these; they’re thin proxy layers.
So I guess the thing that I’m trying to wrap my head around is what are the dimensions on which these different runtimes are competing?
So for AWS, Google or Microsoft, I would say it should be easier to handle or to control these barriers when deploying this kinds of functions. So I would say they are trying to go to some standard, they are trying to make – I would say the same thing, but they are pushing, like “Oh, I’m your friend. We are trying to run together”, but in the end they are trying to compete, because there’s space for everyone. But for me, at some point everyone’s gonna be just in one side, I would say.
Mm-hm. Yeah, if you look at the three that we’re talking about, Kball, it seems that Deno is using – okay, it’s Rust instead of C++… But it’s still V8 at its core. So it’s using more similar things to Node, and really competing on API design, implementation details, the user experience, and performance, and those kinds of things.
In terms of what the competition looks like, it seems like those are the things. And it’s really implementation details then that matter. It’s like, “Well, we implement this using Libuv and you don’t. And Tokyo is better for these reasons, for this use case”, and us users are just kind of like “Okay, I guess I’ll wait and see the benchmarks, or which ecosystem has the packages I need.” That’s a pretty important part of it…
Yeah. And it’s interesting what you’re highlighting with Deno… The differences are not in the code you’re writing so much. I mean, maybe a little bit, but it’s the things around that. And particularly the security - I think I remember that Deno has an ability, similar to what actually the Tauri folks were talking about, where you can say “I only want to expose these system APIs”, and therefore kind of lock down… Even if I’m pulling in npm packages from wherever, I know that if I have not given file system access, it doesn’t matter if somebody inserts some random file read over there, it ain’t gonna happen.
[26:04] Yeah, that is absolutely true. One of the downsides, or one of the bummers about that, Kball, is actually a lot like something you talked about a couple weeks back with tree-shaking; we all got excited about tree-shaking, and then we realized that most of our apps need most of the code, most of the time… And you’re like “Well, we tree-shook it, but not much fell out…”, and that kind of was the reality.
You just give all the permissions, like the apps on the –
Well, with a significantly complicated Deno app, you’re gonna need all those things… And so because it’s at the executable layer, and not package by package, which is just a constrait of the technology today - like, they just actually can’t do it any more granular. There’s people working on that with like shadow realms, that you kind of have to give it most of the things.
So at the end of the day, it’s kind of like “Well, it’s philosophically the opposite of Node”, with regards to security specifically… But in practice - okay, for a very constrained application, then it’s different; but for a broad, general web app or whatever that’s doing lots of things, you kind of have to give it all those things anyway. So that’s a bummer, but it’s just where we are.
You’re gonna get the return of microservices, right? I’m gonna separate my app, not because it makes sense for what I’m trying to do, but because it lets me pinpoint my security pieces…
Right, draw boundaries…
Finally, a good argument for microservices… Nah, just kidding. [laughter]
And actually, there’s a very good post about how James Snell actually wrote the article, speaking about the security and how Deno made… And actually, it’s a very, very close point; like, “Oh, I know you can use a lot of this stuff, but in the end sometimes you want to fix this problem using some infrastructure around this.”
So everything – like, on VPCs, and all of this. It’s rarely when you go to the application. But again, I would say Deno was only able to do this because they were beginning, and they have freedom to do this.
Right. You know, I joke a lot about people who write modern software, because that’s one of the selling points… Like, “This is a modern library.” And really, what that means is you just didn’t start it very long ago. You know, okay, it’s modern because you just started it last week…But actually, Node has been out there, serving millions of users for many years, and continues to do so… And so it can’t actually go modern in certain ways, because it has dependencies. And I’m not talking about npm packages; like, actual real humans depending on it… And that’s the luxury of new things; it’s like “Yeah, we can start fresh, and we can drop all the old stuff…” And we need to do that in order to learn and grow and innovate, but you know, the big dog is still feeding the family, right?
Well, and you notice that as they start to go along, they discover “Hey, some of those things we thought we could drop, people actually care about them.”
Like “Hm… Package management.”
[laughs] Yeah, that’s true.
[30:04] So some things are breaking right now, but we haven’t seen much impact yet, because people are not changed, because maybe – either they are afraid, or it’s close to much of TypeScript; I’ve heard these words, oh my God… [laughter] But yeah, for you who’s listening and you are not using ECMAScript modules, this is crazy nice for all your applications, because now it’s truly asynchronous, now you can share your code without using WebPack, or other mediator to create and make that single culture run in different platforms.
So if we’re gonna go back to the beginning a little bit - and I was gonna ask you, Erick, as an interested developer, if I wanted to code up a Node.js from first principles, from scratch, what would my first few steps be? Now, I know step one could be “Go watch my YouTube video on how to do this”, but assuming that we don’t have that much time and detail, and you’re audio-only, what would be starting point for people who are like “Yeah, I’d like to give it a try, even just to learn.” What would they do?
The first step for me was trying to search examples on GitHub. I found some people trying to use V8, just get a string, parse it, and print out… But it was very hard, because we need the environment. We need to compile the V8, and the doc for me didn’t work well. But I would say for technology or anything, I’m gonna start looking at – right now, I’m gonna go to the GitHub and search if someone is doing something.
But I would say get individual piece… So “Oh, I know right now what V8 can do and cannot do.” We don’t have set timeout, we don’t have set interval, we don’t have set immediate Everything that goes outside, everything that goes asynchronous there, that’s gonna be executed in the future - it’s not the paper of the V8 right now. Everything that is asynchronous in Node is Libuv.
So now, in my case, I ran to the GitHub on the Node.js there, and I noticed that in the beginning Ryan Dahl wasn’t using Libuv. It was libio, I would say, that only worked in Linux and in a few operating systems. That’s why they built the Libuv.
Once you learn the Libuv, and once you learn the role of the V8, you start wondering, “Okay, how can I integrate them?” So you can use Rust, you can use other programming languages, but I would say sometimes we’re gonna feel a bit frustrated; it’s not about the language, it’s not about the APIs. It’s making this whole environment. This is why we’re [unintelligible 00:34:16.20] watch my video. Not because only the content, because I built the whole ecosystem for you on the GitPod there, so you can just start using it right away; you can actually implement challenges, and other stuff. So yeah, learn every piece individually, and try and understand what you could do and you cannot do in every piece.
When you say you built the whole environment on GitPod, real quick… When you say that, what do you mean by that?
Well, I went there, I downloaded V8, and Compile, and got the binary. Libuv - same thing. And now I’m sharing, so everyone can code in the cloud, implement C++ code, and everything from that environment, and you don’t have to worry about enviroment, which I servered at least a month to make this work. So it’s crazy.
[laughs] Erick suffered, so you don’t have to… Just go use his GitPod. Kball?
[35:55] Well, we could go to the ecmascript.com website, in the docs… Like, look what are variables, loops, structures, classes… Promises. Oh, this for me was mind-blowing. For me, promises are async functions that can be used to notify you of something that’s executing in the end. It’s almost like this is not V8, because V8 cannot handle anything in the future. We actually have one which is ‘Kill microtask’’ but it’s so specific that I don’t mention it much. But there, in the end, a promise object is just a wrapper for a callback. So you just have an object that when the C++ calls you back, then you can have that object change. So that object lives on V8.
So as far as I know, I don’t know everything, because I’m still learning a lot… And just a parenthesis here. To make this project work, I actually reached out to the Google team, I reached out to some Node.js developers, and they have no idea how to make this from scratch. So it was crazy. Crazy. So I actually found a guy who worked at the [unintelligible 00:37:12.04] with me, and he used to work at the Libuv and V8. He joined me on a call and helped me out. So I made it work just because of him.
So I would say the limit for V8 is the language itself. I would say the syntax asbtract tree I would say the cons how the interprative variable, how functions it’s being executed. And just it. Just it, It’s a bunch of things but it’s hard.
Right. Does that help, Kball?
It does, and once again, it shines light on how large the Node project is…
Right, that’s what I was saying. Surface area. There’s lots of things that needs it to provide.
Yeah. And for me it was crazy… The V8, when you go to the docs, they have some tutorial there, and they say you can use D8. And it’s like a shell code that you can try using the eval – actually, it’s… I don’t remember anything, but yeah, it’s on the console [unintelligible 00:39:05.13] and I was using Console.log. Console.log doesn’t exist. What?!
So I guess it’s just a sales pitch for this kind of content. I know that a lot of times when we’re looking at continuing education for ourselves, it’s like “How can I learn the next new framework, the next new CSS trick, or whatever it is, to make my apps better?”, that’s all well and good. But also, not just going out, but going down and learning what’s underneath you is really empowering.
However, we should have some stars, right? Because I’m delegating some function, and this function is gonna be started somewhere. But I’m talking about millions of users concurrently accessing that application, and this is being started somewhere. So I was like “Oh my God, so still this is fast, still they know how to handle memory…” It’s so few people that know about V8, and this kind of thing, that usually it’s the same person. When you go to the conferences, when you speak to people at V8, you see that it’s the same 10-20 people that know how the thing works. So this is crazy nice, but it’s still in my opinion, as an instructor right now, everyone should know, everyone should be able to help to improve this technology.
Yeah. The giants are tall, but they’re also few.
And you find out, once you get down there you’re like “Oh, V8. How many people are on the V8 team?” And it’s like 6, or 8, or a dozen, but it’s just not that many peope, and it’s like wow, these 6, 8 or a dozen people pretty much power everything that powers the ecosystem. So it’s just really interesting how we go from the top end side of app developers, and users it’s like millions; and then as you get down to the dependencies, further down, there’s just like one lonely dev in Nebraska maintaining these things, for example…
And for you, who are watching right now - you may be thinking “Oh, maybe ChatGPT knows”, right? No…
[laughs] ChatGPT doesn’t know that.
When you go in there and you try making specific questions about V8, they don’t know; the code doesn’t compile. And it’s hard. Sometimes – I always wondered how those people are maintained. I don’t know, I don’t have this information, but I’m not sure why these docs – are these docs private at Google, or why it’s not so public?
When you go to the docs, you can see the old docs that goes only on the signature. You see what is the input, and the output of the function, but what if you want to convert a function, convert an integer, a string to an integer? In my case - oh my God, I was searching at the tests. So I got to the whole thing I was searching, and said “Oh, maybe I see something how to use”, because of the error that was showing to me. So it’s pretty hard, but I don’t know why this is not so accessible for everyone. Maybe there’s no reason, but I don’t know.
[44:08] Well, speaking of things that ChatGPT doesn’t know… I just broke it yesterday. I asked a question, I was like “How do I do this?” It was like function parameters, blah-blah-blah. And it’s like “Try this.” And I looked at it and I thought “Oh wow, I didn’t know you could do that. That doesn’t look right. But I guess you can do it.” So I went and copied and pasted it, and it broke. And I went back to ChatGPT and like “No, you can’t do that. Here’s what happens.” And it says “Oh, I’m sorry about that. I messed up. Here, try this.” And it gave me the exact same code. I’m like “Ohh, I’ve found a loop here. It’s just gonna keep telling me the exact same code, even though it’s broken.”
ChatGPT is an amazing technology in an even more amazing hypecycle.
Yes. The hypecycle is strong right now.
It can do a lot of really powerful and cool things, but it ain’t AGI.
No. And we humans are very good at cherry-picking the best examples and putting them on social networks, so that we get lots of retweets and likes.
[laughs] But for me – I’m curious, right? For everything. Like, oh, why is my webcam green now? I try to make a lot of questions. And sometimes these questions – it’s like 1 AM, so I go to ChatGPT, how to make a web driver from scratch. Right now I’m trying to do the same thing I did for Node, recreating Node, but I want to do on the mobile devices. So I was like “Maybe the idea is the same, but don’t they use Libuv to make this stuff?” I don’t know, the mobile seems not so powerful in processing power, and this kind of stuff… And in the end, I was searching there… Every of those platforms, they are using C++, or bytecode to make those bridges… So you can call native APIs from Android, from iOS, the same way we are doing on the desktop. So this is how native scripts, React and Native, they are doing behind the scenes. But again, I already like a week worked on this and I couldn’t set up the environment. This is so hard…
Oh, even harder, probably… Yeah. So that’s your next video, is gonna be how to recreate React Native from scratch, or what’s it gonna be?
Well, I wouldn’t say the next, because I’ve already been spending a lot of time on this…
…but for these conferences I’m going to speak about recreating Node in a lot of conferences, and I would put it as like a plus. “Oh, look, you know how to make an input on your Android. This is how we can do the same idea.” But it’s a bit harder, because in my case I don’t know Android, and I don’t know iOS. And I’m not an expert on C++. So you can see a lot of bad practices on my code, because I’m just – some people be like, try and find the answeres. So this is a bit harder for me, but I don’t give up easy.
I feel like that’s actually a key to becoming a good developer, and to going deep in all of these things.
It’s like, once you get outside of the domain you know, it’s gonna be hard. There’s gonna be dead ends, there’s weird, esoteric knowledge that somehow has not made its way into the documents, or into ChatGPT, or into whatever else you’re using to teach yourself. And you’re gonna get stuck for a while. And you’ve gotta not give up to easy, and keep banging your head, and try changing your question or your search phrase until you figure out what is the weird thing that somebody’s documented in a blog post that wasn’t picked up by whatever, that gets you past that barrier.
It’s kind of like a video game… There’s two kinds of video game players. There’s the ones who get stuck and they’re like “This isn’t fun anymore. I’m here to have fun. I’m gonna move on.” And I totally understand that, because that’s logical. Like, if the goal is fun, and this isn’t fun - why am I doing it? But then there’s the kind of video game player that I am, where it’s like “Oh, I’m stuck right here. My life is gonna be hell for the next seven hours until I get past this level.” And I’m just gonna go over and over and over until I get past it.
[48:08] Sometimes I feel like programming has been like that, for me personally, where I’m like “I’m just stuck. I don’t know my way out.” This is pre-GPT tooling, pre-lots of helps… And I’m like “Well, it’s me and the keyboard here. Let’s bang our head on it until we finally figure it out.” And it sounds like you’ve persevered through some of that yourself, Erick, in getting this thing out there.
Right?! Well, I guess as a fellow content creator then, let me ask you a question about that. I guess all three of us here are people that create stuff for the internet… You had this video - we’ll put it in the show notes; it’s the one that we’ve been talking about the whole show. It’s an hour and 25 minutes, it has chapters and stuff, it has all the things that a polished video should have. And literally, months of work, literally, before you could even start recording. When you’re making something like that, do you ever just stop and say “What if nobody cares? What if nobody watches this?” I mean, you’ve done a great job; it has 123,000 views, so - massive success on that front. So that’s amazing. It’s gotta feel good. But do you ever wonder, like “What if I put –” Because a lot of us do put things out into the world and it’s like “Meh…” And you’re like, “Dang…”
It happens almost everytime with me. But I noticed sometimes I try creating – actually, right now I’m more trying to balance… This is content that I love to do; this is something that I really like doing. I know that’s not gonna be hype. This is actually my story here, producing content in Brazil. Everyone else is trying to teach beginners and intermediate developers… But I’m seeing a limbo. So from the intermediate to advanced, and especially, there’s no one speaking… Because it’s hard to put people in my marketing team to say “No, these guys didn’t buy. Those people don’t buy courses I was like “Let’s try out.” I wanted to check this out, I wanted to fail, but I wanted to fail fast.
But in this case of the recreating Node, it was something so interesting for me that even if it was only me, and I was like “Oh, mom, come here and see what I did!”
This is like your opening line in a bar?
Right. Well, that’s the right attitude right there.
[51:57] Do you feel like you have a sense of when something’s gonna go or not? Because I was just talking to somebody about this, and they’re like “Yeah, I had one that went really well. So now can I just take that?” I’m like “You know, every single thing that I put out there, I don’t know if it’s gonna go or not.”
And there have been things where I was like “This is gonna go, this is gonna go”, and it’s like… Nothing. And then I had things where I threw this out in response to somebody’s question, and it took me an hour to write up in more detail, and I threw it up there, and it’s still generating hundreds of page hits off of Google every day. And I’m just like “Really? That’s what’s going?”
I have a formula right now, that is every day I publish a text - it’s like 2,000 characters on LinkedIn, and this is, I would say, like my school. So I publish there, and if it goes viral, like a thousand likes or more, I know this was a validated content. So from there, I actually push forward and I create a YouTube video. So when this goes viral on YouTube, I get this and I publish a blog post and create a talk to submit at [unintelligible 00:53:07.05]
So the whole point for me is trying to fail fast. But as you said, sometimes – I’m really tired sometimes. We cannot be creative every day. So sometimes I’m like “Oh, I’m gonna just say whatever I had from some experience. “Five tips I suffer, you shouldn’t suffer.” This goes high. I was like “Oh my God.” I didn’t put enough energy there, and it goes viral. So sometimes it goes as we don’t expect. I would say the constancy, and publishing every day is what makes our life easier.
Right. We can’t control the outputs… Like, we can’t control the results, but can control our inputs or our own personal outputs. So focus on both quality and consistency has been my formula… More than quantity. I mean, consistent with a quantity is a good thing… But just focus on doing things that you think are good, and have a thick skin when the thing that you really thought was awesome goes out with a dud… And just be genuinely happily surprised when things do manage to land. That seems to carry is through…
But it can hurt your feelings a little bit when you’ve put a lot of work into something and it doesn’t seem to do what you were hoping it does. I don’t know if “hurt your feelings” is the way to – just disappoint, I guess.
Okay, we’re coming up at the end here… Any final thoughts, Erick, or Kball? Anything else we wanna talk about today before we call it a show?
Oh, just about this last piece… If you are thinking to build content, sometimes take a break. Don’t push too much at the beginning, because it’s so stressful. You don’t make enough money sometimes; you get a lot of haters… As you are growing, and you have a lot of people saying “Oh, this is wrong, this is wrong, this is wrong”, but you can see that the amount of the haters is very low compared to the people that you actually help. So educating people, and actually sharing what’s in your mind would help a lot of other people. So this is really nice.
Well said. Kball. Last chance to say something before the show is over.
I love what Erick just said. Our tendency as human beings is to focus on the bad. So whether that’s the piece that didn’t go, that you just published and it was dead silence, or the one hater who’s out there saying “This is wrong. This is terrible.”
I think all of us - that’s where our emotions get caught up, and it’s easy to overlook all the people who read it and were like “This is great” and never said anything. But there’s a lot of those out there. So yeah… And I think it comes back to the same thing you were talking about with development, right? Like, push through. Be stubborn.
[55:54] Dang it, I’m publishing for me as much as for anyone. And you never know… This is the last thing - you never know where someone impacted. I had somebody reach out to me a week or two ago, that was like – he’s an entrepreneur, he’s had multiple companies, and he said “I don’t know if you remember me, but I share your blog post from a decade ago with every company that I work with. Would you like to reconnect?” And I was mind-blown. I didn’t even know if that blog post is still alive. It was on an old site… It’s that old. I had no idea. And you have no idea which of the things that you wrote changed somebody’s trajectory, and they are going off – or you recorded in YouTube, or whatever it is. But it does happen. People see your stuff and it can change their lives. So don’t stop writing, or publishing, or whatever it is because of some hater.
Yeah. Just to pile onto that… I would say celebrate your wins, no matter how small, even in development. You get past that one thing, celebrate that win, whatever that looks like for you. Maybe it’s just a step back and play your favorite song, or maybe you’re gonna go out for a beer, or ice cream, or maybe you’re gonna take a day off… Whatever your gonna do. But celebrate those wins, because they really do power you through the downward parts, the struggle. It is toil, both when we’re talking about software development and content creation, and really anything that you do in life that’s worthwhile, it’s gonna have some toil… And so if you can celebrate those wins, the big one, the small ones, then you can kind of get through the doldrums on the other side.
Speaking of doldrums… We’re coming to the end of our show; it’s beginning to tail off. Erick, where is the best place to connect with you on the internet?
Well, I would say follow me on LinkedIn. LinkedIn is the best one. But if you want to chat directly with me, maybe Twitter, because I reached the friendship amount there, and I have no idea how to delete, and LinkedIn is so messy for this… But Twitter, or Instagram… I’m very accessible, but I’m busy as well, so don’t be sad if I take like a week to respond, because it’s been crazy; you see, the content is crazy… But I always try to give some advice. And actually, sometimes I don’t know the answer, but I know how to link to other people as well.
Very good. And if you want to see Erick in the flesh and you happen to be in and around Brussels in May, check out BeJS Conf. One last shout-out to our friend Omar, and that conference, which I’ve heard only good things about, which as a conference organizer I know feels good… Because they have the same struggles as content creators have, as well as that one hater who leaves the bad review… Don’t be that guy or gal; don’t be that one hater.
Anyways, let’s call it a show. Erick, I appreciate you coming on. Kball, I always appreciate you. I’m Jerod, this is JS Party, and we’ll talk to you on the next one.
Our transcripts are open source on GitHub. Improvements are welcome. 💚